Rechargeable, single-cell batteries, such as Li-ion/Li-polymer batteries, are becoming commonplace DC power supply cells for a variety of portable and handheld products. As one would expect, the demand for increased functionality and longer run time of such battery-powered products has resulted in a demand for increased battery cell capacity, with an attendant increase in power required to charge them. A typical single-cell battery charger controller has a relatively compact and portable arrangement, as diagrammatically illustrated in FIG. 1. As shown therein, the charger system includes an external power adapter 10, having an external power pair that is configured to be plugged into a source of external power, such as a 110 VAC wall outlet or automobile electrical system, and a power cable connector 12 that mates with a connector receptacle 22 of a DC-DC converter/charging unit 20. The charger controller unit proper is configured to maintain a battery 30 to be charged.
At present, the majority of DC-DC converter/charging units of such portable battery charger arrangements are based upon a linear transfer function design, such as that diagrammatically shown in FIG. 2. In a linear charger, the wall adapter 10 serves as a DC voltage source and typically has a substantially steady input voltage vs. current characteristic shown in FIG. 3. The charger controller's input voltage as sourced by the adapter 10 may be slightly higher than the nominal (floating) voltage of the battery to be charged, and remains effectively constant over an operating current range set by the charger controller.
The output of the adapter 10 is coupled to a controlled current flow path circuit, such as, but not limited to, a bipolar transistor or MOSFET, shown at 21, the source-drain path through which current flows from the adapter 10 to the battery 30 being charged. A control circuit 25 for controlling the operation of the current flow path circuit 21 has a current sense link 26 (which may be a sense, resistor) which monitors the current through the current flow path circuit 21, as well as a voltage sense link 27 coupled to monitor the voltage of battery 30 as it is charged. The control circuit 25 typically comprises conventionally employed threshold sensor and comparator-based control components of the type used in a variety of current, voltage, and switching control applications.
The operation of the linear charger of FIG. 2 may be readily explained with reference to the waveforms shown in FIGS. 4, 5 and 6. At the beginning of the charging cycle, the battery voltage shown at VBAT in FIG. 4 is at some less-than-nominal value, VBAT0. With the MOSFET 21 being rendered conductive by control circuit 25, a prescribed constant charging current ICHG flows through the MOSFET's source-drain path from the adapter 10 and into battery 30. As shown in FIG. 5, this regulated charging current continues to flow up to the point at which the battery voltage reaches its floating (nominal) voltage VBATNOM. Once the battery voltage reaches its nominal voltage, the control circuit 25 regulates the battery voltage at this target value, causing the current flowing in the MOSFET 21 to slowly decrease until completion of the charge. As will be appreciated from the foregoing description, and as shown in FIGS. 4 and 5 in particular, a typical linear battery charger exhibits a constant current (FIG. 5)-constant voltage (FIG. 4) charge profile.
In order to match an increase in cell capacity, the charging current needs to increase. However, as shown in FIG. 6, it suffers from substantial thermal dissipation, due to higher charging current. In particular, at the beginning of a recharging cycle a ‘fully’ discharged battery may exhibit a voltage on the order of 2.5 VDC, and a typical floating voltage value is on the order of 4.2 VDC. If, for example, the input voltage is selected to be 5.0 VDC (which is only 800 MV above the 4.2 V floating voltage) and the battery charging current is one ampere, the thermal dissipation will be (5 V−2.5 V)×1A=2.5 W at the beginning of the charging cycle.
One approach to reduce the thermal dissipation is to employ a pulse charger, such as that illustrated in FIG. 7, which is similar to the linear charger of FIG. 2, except that there is no current sense link, the current limiting function being built into the adapter, as shown by the voltage vs. current relationship of FIG. 8. The operation of a pulse charger may be understood by reference to the diagrams of FIGS. 9, 10 and 11. During constant current mode (FIG. 9), the control circuit 25 fully turns on the current flow/pass element (MOSFET) 21. As a result, the voltage across the pass element will be either a saturation voltage (if element 21 is a bipolar transistor) or, in the FIG. 7 example of using a MOSFET, will be the product of the charging current and ON resistance RON of MOSFET 21.
As shown in FIG. 8, the adapter 10 operates in a constant current region and its output voltages collapses to a voltage slightly higher than the battery voltage. Thus, the charger does not need to control the charging current, which is limited by the adapter (the charging current source). The thermal dissipation associated with a pulse type of charger is the product of the voltage across the pass element 21 and the charging current. For example, if the charging current is one ampere, as in the linear case, described above, and the ON resistance RON of the pass element (MOSFET) is 300 milliohms, then the power dissipation will be 0.3 Ohm×1A×1A=300 mW, a much smaller value than 2.5 W for the case of a linear charger described above.
As shown in FIG. 10, as the battery voltage approaches the floating or nominally fully charged battery voltage, the pulse charger starts to alternately turn the pass element (MOSFET 21) on and off, and gradually reduces the duty ratio of the ON time, until termination of the charging cycle. Power dissipation (shown in FIG. 11) is 300 mW when the pass element is on and zero when it is off. Therefore, the average dissipation is less than 300 mW during the pulse phase.
Although low power dissipation is a principal advantage of a pulse charger, such a charger requires a particular type of adapter—i.e., a current-limiting adapter. The main disadvantage of a pulse charger is the fact that, during pulse mode operation, it produces pulsed voltages at both the input and output of the charger, which constitute potential electromagnetic interference (EMI) noise that may affect the operation of one or more electronic circuits in the device powered by the battery being charged. In addition, the pulse charger may affect the lifetime of the battery and is not recommended by most battery cell manufacturers.
A third type of charger that may be employed is a switching charger. A switching type charger requires more components (including a bulky output inductor) and switches large currents at high speeds, making it the most noisy and complicated among the three types of chargers. It is most practical for high-current applications, such as notebook computers.